Re: 1b (or 1aii because of my love for indenting): That makes sense. I think I agree with you, and I’m generally unsure how much of a factor what I described even is.
Re 2: Yeah, this seems right. I do think some of the selection effects might mean that we should expect that forecasting is less promising than one might think given excitement about them in EA, though?
Re: 3: Thanks for clarifying! I was indeed not narrowing things down to Tetlock-style judgmental forecasting. I agree that it’s interesting that judgement-style forecasting doesn’t seem to get used as much even in fields that do use forecasting (although I don’t know what the most common approaches to forecasting in different fields actually are, so I don’t know how far off they are from Tetlock-style things).
Also, this is mostly an aside (it’s a return to the overall topic, rather than being specific to your reply), but have you seen this report/post?
Re: 1b (or 1aii because of my love for indenting): That makes sense. I think I agree with you, and I’m generally unsure how much of a factor what I described even is.
Re 2: Yeah, this seems right. I do think some of the selection effects might mean that we should expect that forecasting is less promising than one might think given excitement about them in EA, though?
Re: 3: Thanks for clarifying! I was indeed not narrowing things down to Tetlock-style judgmental forecasting. I agree that it’s interesting that judgement-style forecasting doesn’t seem to get used as much even in fields that do use forecasting (although I don’t know what the most common approaches to forecasting in different fields actually are, so I don’t know how far off they are from Tetlock-style things).
Also, this is mostly an aside (it’s a return to the overall topic, rather than being specific to your reply), but have you seen this report/post?